Unformatted text preview: Second Edition 915 where c = 2 / 2 . The derivative (with respect to c) of this last expression is bfn1 (bc)  afn1 (ac), and hence is equal to zero if both c = 1 (so the interval is unbiased) and bfn1 (b) = afn1 (a). From the form of the chi squared pdf, this latter condition is equivalent to fn+1 (b) = fn+1 (a). d. By construction, the interval will be 1  equaltailed. 9.53 a. E [blength(C)  IC ()] = 2cb  P (Z c), where Z n(0, 1). b. [2cb  P (Z c)] = 2b  2 1 ec /2 . 2 2 c. If b > 1/ 2 the derivative is always positive since ec /2 < 1.
d dc
2 9.55 E[L((,), C)] = E [L((,), C)S < K] P (S < K) + E [L((,), C)S > K] P (S > K) = E L((,), C )S < K P (S < K) + E [L((,), C)S > K] P (S > K) = R L((,), C ) + E [L((,), C)S > K] P (S > K), where the last equality follows because C = if S > K. The conditional expectation in the second term is bounded by E [L((,), C)S > K] = = > = E [blength(C)  IC ()S > K] E [2bcS  IC ()S > K] E [2bcK  1S > K] (since S > K and IC 1) 2bcK  1, which is positive if K > 1/2bc. For those values of K, C dominates C. 9.57 a. The distribution of Xn+1  X is n[0, 2 (1 + 1/n)], so P Xn+1 X z/2 1 + 1/n = P (Z z/2 ) = 1  . b. p percent of the normal population is in the interval zp/2 , so x k is a 1  tolerance interval if P ( zp/2 X k) = P (X  k  zp/2 and X + k + zp/2 ) 1  . This can be attained by requiring P (X  k  zp/2 ) = /2 and P (X + k + zp/2 ) = /2, which is attained for k = zp/2 + z/2 / n. c. From part (a), (Xn+1  X)/(S 1 + 1/n) tn1 , so a 1  prediction interval is X tn1,/2 S 1 + 1/n. ...
View
Full
Document
This note was uploaded on 02/03/2012 for the course STA 1014 taught by Professor Dr.hackney during the Spring '12 term at UNF.
 Spring '12
 Dr.Hackney
 Statistics

Click to edit the document details